RVF LLMInstall

Summary

Generate images using a selected provider, model, message prompt, and optional settings.

Inputs

• provider (required) — Select an AI provider that supports image generation.
• model (required) — Select a model from the chosen provider that can generate images.
• messages (required) — Provide an array of message objects (role and content) that form the prompt.
• options — Configure generation parameters such as temperature, max tokens, and enable built‑in tools.
• builtInTools — Enable or disable built‑in tools like web search, file search, or code interpreter.

Output shape

an array of objects representing the generated image response

Each array element contains a json field with the API response; errors are thrown unless continueOnFail is enabled.

Examples

Example 1: Create an image of a futuristic cityscape

provider: 'midjourney', model: 'midjourney-v5', messages: [{role:'user',content:'A futuristic cityscape at sunset'}], options: {temperature:0.7, maxTokens:1200}

Example 2: Generate an image with web search enabled

provider: 'gpt4-image', model: 'gpt4-image', messages: [{role:'user',content:'Current stock price of TSLA'}], options: {webSearch:true, temperature:0.5}